Catalytic and non-catalytic mechanisms of histone H4 lysine 20 methyltransferase SUV420H1

Abstract

The intricate regulation of chromatin plays a key role in controlling genome architecture and accessibility. Histone lysine methyltransferases regulate chromatin by catalyzing the methylation of specific histone residues but are also hypothesized to have equally important non-catalytic roles. SUV420H1 di- and tri-methylates histone H4 lysine 20 (H4K20me2/me3) and plays crucial roles in DNA replication, repair, and heterochromatin formation, and is dysregulated in several cancers. Many of these processes were linked to its catalytic activity. However, deletion and inhibition of SUV420H1 have shown distinct phenotypes suggesting the enzyme likely has uncharacterized non-catalytic activities. To characterize the catalytic and non-catalytic mechanisms SUV420H1 uses to modify chromatin, we determined cryo- EM structures of SUV420H1 complexes with nucleosomes containing histone H2A or its variant H2A.Z. Our structural, biochemical, biophysical, and cellular analyses reveal how both SUV420H1 recognizes its substrate and H2A.Z stimulates its activity, and show that SUV420H1 binding to nucleosomes causes a dramatic detachment of nucleosomal DNA from histone octamer. We hypothesize that this detachment increases DNA accessibility to large macromolecular complexes, a prerequisite for DNA replication and repair. We also show that SUV420H1 can promote chromatin condensates, another non-catalytic role that we speculate is needed for its heterochromatin functions. Together, our studies uncover and characterize the catalytic and non-catalytic mechanisms of SUV420H1, a key histone methyltransferase that plays an essential role in genomic stability.

Figure 1.. Activity and structure of SUV420H1-nucleosome complex

A. Catalytic activity of SUV420H1 on H2A and H2A.Z H4K_C20me1 nucleosomes measured using an endpoint methyltransferase assay. Values above the bars represent the ratios between SUV420H1 activity measured on H2A.Z and H2A nucleosomes. B. Michaelis–Menten titrations of H4K_C20me1 nucleosomes containing H2A and H2A.Z with SUV420H1. The K_M and k_cat values of the fitted data are reported in the graph. Each data point represents the mean ± SD from experiments repeated at least three times. C. Structures of SUV420H1-H2A.Z nucleosome complex shown in two different views. The color scheme for complex composition is indicated below the model.

Figure 2.. Interactions of SUV420H1 with H2A.Z nucleosome

A. Overview of the contacts between SUV420H1 and H2A.Z nucleosome. B. Close-up view of interactions between SUV420H1, histone H4, and DNA. On the right is a representative methyltransferase activity of wildtype SUV420H1 and R286A mutant on H2A and H2A.Z nucleosome. C. Detailed interactions of histone H4 with the SET domain of SUV420H1 with representative methyltransferase activity of wildtype SUV420H1 and mutants (E314K and S251A). D. Detailed interaction between SUV420H1 and acidic patch residues of H2A.Z nucleosome with representative methyltransferase activity showing the effect of mutants (R352A and R357A). For each methyltransferase assay, data points and error bars represent the mean ± SD from three independent experiments.

Figure 3.. Distinct cryo-EM reconstructions of SUV420H1 bound to H2A.Z nucleosome

A. SUV420H1 bound to one face of the H2A.Z nucleosome. B. SUV420H1 bound to both faces of the H2A.Z nucleosome. C. H2A.Z nucleosome alone. D. Cartoon depiction showing how detachment of DNA correlates with SUV420H1 binding occupancy.

Figure 4.. SUV420H1 C-terminus and nucleosome interactions

A. Diagram of SUV420H1 highlighting the C-terminal region colored in orange. B. Close-up view showing the cryo-EM density of SUV420H1 C-terminus, histones H3 and H2A.Z. The interacting region between SUV420H1 C-terminus and aN helix of histone H3 is circled in white. C. Circular diagram showing intermolecular crosslinks found in the SUV420H1-H2A.Z nucleosome complex. The crosslinks between SUV420H1 C-terminus and aN helix of histone H3 are shown in red, and all other crosslinks are shown in grey lines. The intramolecular crosslinks are not presented for simplicity. D. Cryo-EM reconstruction of SUV420H1 C-terminal peptide bound to H2A.Z nucleosome, showing DNA detachment.

Figure 5.. Biophysical and chromatin condensation activity of SUV420H1

A. Schematic representation of the optical tweezers setup where a doubly biotinylated 12-mer nucleosome array with long DNA handles is tethered to a pair of streptavidin-coated beads. B. A representative force-extension curve for an unbound histone H2A 12-mer array (black) and one for a SUV420H1-bound array (purple). C. Distribution of transition forces observed in the force-extension curves in (B). D. Micrograph panel of H3K9me3 chromatin array mixed with serial dilutions of SUV420H1 showing chromatin condensation. E. Fluorescence microscopy panel of H3K9me3 chromatin array (red) with HP1α (cyan) and serial dilutions of SUV420H1

Figure 6.. Cellular activity of SUV420H1

A. Representative western blot showing the level of H4K20me2/3 in Suv420H1 MEFs, loading control αTubulin. B. Relative abundance of histone H4 lysine 20 methylation in wildtype, R286A and ΔC-terminus Suv420H1 containing MEFs. Bars graphs represent the means ± SEMs of four independent biological replicates. A 2way ANOVA statistical analysis was performed (*p < 0.05, ***p < 0.001, ****p < 0.0001) C. Percentage of cells with >3 53BP1 foci per nucleus in SUV420H1 knockout, wildtype, R286A and ΔC-terminus MEFs cells. Bars graphs represent the means ± SEMs of three independent biological replicates. A 2way ANOVA statistical analysis was performed (*p < 0.05, ****p < 0.0001)